Continuing from Part I
Baseball Prospectus: One risk that you took last year was an unorthodox relief strategy. In retrospect, do you think it would’ve been more prudent to just go ahead and try it, rather than announcing it beforehand?
Epstein: Well, we didn’t actually announce it. I think we were too open in responding to questions about it. There’s a difference. I know it’s semantics but it’s a difference nonetheless. I will say this: One of our biggest mistakes of last season, and something we’ve corrected, is that we were just far too open in general. We had too big a circle of information here, too many people who knew what was going on. Word eventually leaked out because we were too open about our plans, and I think it hurt us. That’s a mistake I think we’ve learned from, and we’ve really tightened up our circle of information. We’re playing our cards closer to the vest, and I think that was demonstrated with the Schilling trade, because that was a big deal that caught, I think, the whole world by surprise at the time it was announced. Obviously we didn’t do as good a job when the potential A-Rod trade was discussed so nationally, but that certainly didn’t come out of Boston–that came from somewhere else.
In retrospect, if I could do last year’s bullpen over again, for starters I hope I’d do a better job putting together quality relievers. I didn’t do a very good job. No matter what we said about it, I didn’t have the right guys in here to get the job done.
I think both sides of the debate missed a big issue. The general public and traditional media thought we were trying to do a “bullpen by committee,” a revolutionary idea. They decided to just blame the whole thing on Bill James, and got it all wrong. On the other side, the new school guys like yourself thought, great, they’re not going to overpay for saves, and they’re going to try to apply what Bill wrote about the ace reliever and unconventional usage to create the “optimal bullpen.” The truth was really somewhere in the middle.
In a vacuum, I’d love to try to create optimal bullpen usage and create a small competitive advantage–small–by doing that. But that’s in a vacuum. I don’t think this is the right time or place to try it, and we didn’t have the right personnel to try it. We didn’t have a dominant reliever. In reality, what happened was we made a conscious decision not to overpay for Ugie Urbina. And we were trying to trade Shea Hillenbrand for Byung-Hyun Kim. But it didn’t work (at the time); we couldn’t get it done during the off-season. We came close, but they wanted a prospect. We thought eventually it would get done, and it did get done–in May. So we basically had a bullpen without a dominant reliever, and that’s not going to work too often. And it didn’t work.
BP: So now you’ve got your dominant reliever.
Epstein: I think we have several now. I hope we have several. But yeah, certainly Keith Foulke should be dominant.
BP: Will we see Foulke trade 15 easy save opportunities for 15 tie-game, higher-leverage appearances?
Epstein: Maybe. I think where we stand right now is that the best way to try to improve bullpen usage is to try to have your best pitcher throw the most important innings possible. It’s going to be really hard to say “OK, we’re going to use him in the seventh and eighth inning and let someone else pitch the ninth.” What we’re going to do is be very aggressive using him in the eighth inning in close, big situations–tie games and one-run games–and let him pitch both the eighth and ninth. Hopefully he’ll pitch more innings. We’ll be aggressive using him in tie games, and aggressive using him for multiple innings. And that will shift his usage enough that I hope it will be improved.
BP: Last year Todd Walker, Trot Nixon and David Ortiz all logged a number of ABs against lefties over the course of the season despite a historical inability to hit same-side pitching. With a hand-picked manager in the fold, do you think you’ll be exploiting platoon match-ups more often in the future?
Epstein: That’s too complex a question to answer with a blanket yes or no. We believe in putting our players in the best possible position to succeed, and we believe in track records. And I believe in platoon splits; we still have a debate internally, and I think it’s debated in the sabermetric community, whether platoon splits are unique to the player or whether there is sort of a standard split and everyone deviates around that mean split, the average split. Nonetheless I believe in them, and I believe in Earl Weaver. I think you can build a really effective platoon if you have the personnel. I think you’ll see that we have some effective possible platoons out there. Terry Francona is very aware of his players’ strengths and weaknesses, and he’ll do everything he can to put them in the best possible position to succeed.
Oh, and I want to add this too: Of all those things I said I believe in, I also believe in development. I also believe, and we studied the issue and it’s hard to determine whether or not there’s an actual statistical ability to do this, but I think it is true that for a player to show he can hit lefties, you have to give a player a chance to hit lefties. Boston in 2003 and 2004 is not the optimal time to try that, because we’re trying to win every possible game that we can. But I was in San Diego when Ryan Klesko was given his opportunity to play every day, and he blossomed into a better hitter and started to hit lefties a little bit better. So there’s something to be said for development, and opportunity, and improving on your weaknesses. (Laughing.) But I probably believe in platoon splits a little more than I believe in that.
BP: You’ve assembled a pitching staff that is heavy on strikeouts and fly balls, save Derek Lowe. With that in mind, why sign Pokey Reese to play second base? Doesn’t a good glove at second have less value to the Red Sox than other characteristics?
Epstein: In an ideal world, we’d have a pitching staff heavy on strikeouts and ground balls and light on walks and home runs. I think what we have is a pitching staff heavy on strikeouts and light on walks, which is something that you really want. We did add a couple of pitchers who are extreme fly ball pitchers, but that was unintentional and came as a byproduct of quality. That’s where the quality was. Schilling and Foulke, in our minds, were two of the top three or four available pitchers in the whole game. We weren’t going to force-feed ground ball pitchers. The numbers aren’t dramatic enough to make Pokey’s defense insignificant to this team. And with certain pitchers it gives us the opportunity to put a pretty good on-base guy in Mark Bellhorn in the lineup as well. I think there’s enough of a balance between ground ball pitchers like Lowe, Timlin, and Mendoza with the rest of our staff to get the right players out there.
But you’re right, it’s not ideal. If you had tremendous infield defense all around, you’d want as many groundball pitchers as possible.
(Note: It came as news to Epstein that his staff skewed toward fly balls, and shortly after the interview he submitted some “quick and dirty” research to Baseball Prospectus via e-mail.)
Using our 10 pitchers with established ML track records (this leaves out Arroyo, a modest fly ball guy) and using their 2002 and 2003 seasons, our 2004 staff has a GB-FB ratio of 1.46 versus the ML average in that span of 1.21. Over the last two seasons, the average ML pitcher has a K/BF (strikeouts per batters faced) rate of 16.6%, a GB/BF (ground ball out per batter faced) rate of 31.8%, and a FB/BF (fly ball out per batter faced) rate of 26.1%. Our projected staff had a K/BF rate of 22.3%, a GB/BF rate of 33.7%, and a FB/BF rate of 23.0%.
In reality, our 2004 staff should produce more strikeouts and groundball outs, but fewer fly outs, than the average staff. Derek Lowe has a lot to do with the high GB rate, but it exists nonetheless.
BP: Virtually no Sox batter can be expected to be better in ’04 than they were in ’03, and…
Epstein: Uh, I don’t know if that’s quite true, but go ahead.
BP: Well, can be expected to. It does seem like they outperformed…
Epstein: Virtually no batter? I can name a couple who should be expected to be a little bit better.
BP: Name them.
Epstein: I think Kevin Millar will be better in ’04 than he was in ’03. I think Pokey Reese will be better–not better than Walker, but better in ’04 than ’03. And I think Nomar Garciaparra will be better in ’04 than he was in ’03. That said, we are, in our internal planning, not expecting to score quite as many runs as we did last year. I think it would be unrealistic to expect that out of us. But we’re OK with that, because we’re going to score a number of runs we’re comfortable with and we’re going to allow a lot fewer runs.
BP: Considering the Red Sox resource pool in contrast to a flat free agent market, does investing in the farm system make any sense rather than just reloading with free agents?
Epstein: I think there’s a balance. Over the long haul there is just too much of a positive impact of having a strong farm system to ignore it. The amount of money we spend on the farm system is small when compared to the size of our payroll, so I think taking money away from the farm system and putting it into free agency would not be wise. So many good things happen when you can develop an abundance of prospects on a yearly basis. You can handpick the ones you want to keep based on ability, position and makeup, and you can move the ones you want for help at the big league level. And the ones you want to keep, you might have an inexpensive solution for up to six years. That’s too valuable to turn your back on, and in fact it’s something we’re emphasizing. And the only way to get a player who has grown up practicing the “Red Sox way,” so to speak, reflecting the Orioles way of 20 years ago, is to do it yourself. If you want a player who cares about the things we care about, both on and off the field, you have to develop him yourself.
BP: A few quick ones, you can answer these yes or no if you’d like. SABR-L, the listserv for the Society for American Baseball Research, recently finished one of its semi-annual debates about “clutch” hitting. Does clutch hitting exist?
Epstein: We have this debate a lot too. I guess I’ll cop out and say it doesn’t exist, but there are certain guys I want up, OK. Clutch hitting does not exist, statistically. I put a lot of weight on that. But I can think of two players with really similar performance on a certain club where, as a general manager, there is one guy I would want up with the game on the line and one that I wouldn’t. So you can make of that what you will. Technically it does not exist, but at the same time the human element does come into play and can easily confuse one into thinking it does exist.
BP: How about the “hot hand”?
Epstein: No. I think regression to the mean is a more powerful force than the “hot hand.” But an element in baseball overlooked by sabermetricians is coaching, instruction, development, improvement. There is such a thing as a player, even mid-season or mid-game, making an adjustment and changing certain elements of his performance. So no, I think the “hot hand” doesn’t necessarily exist, and regression to the mean is powerful. But players are dynamic–they’re not bound by their track records.
BP: How about “the curse”?
Epstein: I don’t think that exists. And we look forward to proving it.
Joe Sheehan, Dayn Perry, and James Click contributed questions for this interview.
Nathan Fox is a freelance writer living in Boston. He can be reached at firstname.lastname@example.org.